'It's the Demography Stupid': Ethnic Change and Opposition to Immigration
Eric Kaufmann, Birkbeck College, University of London
In May 2014, as the economy looked to be recovering, immigration re-emerged as the top
concern of British voters. I echo Matt Goodwin (see article in this volume) in contending that
cultural change is what principally animates anti-immigration voters. The issue therefore
cannot be defused with economic or political palliatives. Political entrepreneurs like Nigel
Farage are profiting from low trust in politicians, but neither this, nor an aggressive tabloid
media, adequately explain the importance of this issue in the public mind. Instead, I point to a
slow-moving, unglamorous but highly consequential source of discontent: ethnic change. In
England, over the past decade, ethnic change has outpaced the coping mechanisms of the
ethnic majority.
This paper urges policymakers to take a long view based on history, and use a
comparative, wide-angle lens on the politics of immigration. Ethnic change is historically
associated with calls for immigration restriction and a general mood of defensive ethnic
nationalism. This has been true in England, Scotland, western Europe and North America in
the nineteenth and twentieth centuries. Demography is not destiny: integrating shocks such as
wars or major ideological changes create new frames which may render increased diversity
palatable - as in present-day Canada or the post-bellum United States, or minimal diversity
unpaltable, as in Nazi Germany. Yet these historical junctures are the exception and not the
rule. Neither are present in contemporary England. In the rest of Europe it is only in
Germany, for historical reasons, and perhaps Sweden, that a pro-immigration ideology serves
as a symbol of national uniqueness. Elsewhere in Europe, including England, ideological
constraints are more muted. This means the most important driver of majority attitudes is
demography: the balance between ethnic change and integration.
Ethnic change in England would continue even if immigration ground to a halt. The
doubling in England's minority share (defined here as black an minority ethnic) from 6 to 12
percent in the 2000s was powered as much by age structure and fertility trends as
immigration. This makes it imperative to examine the forces of integration which may help
reduce the tension generated by ethno-demographic change. There is, additionally, an
important difference between European and minority (i.e. nonwhite) diversity in that the
children of European immigrants are more readily accepted through assimilation into the
majority group than the latter. This suggests that UKIP's call to end the free movement of
Europeans into Britain and replace them with skilled migrants from elsewhere would not
necessarily reduce anti-immigration sentiment or far right voting.
A Note on Method
This research is based on an ESRC-funded project, "Exit, Voice or Accommodation?:
white working-class responses to ethnic change in Britain", which I hold jointly with Gareth
Harris of Birkbeck College.1 Findings are based on quantitative analysis of several large
datasets, including the Citizenship Surveys, Understanding Society (UKHLS), British
Household Panel Survey (BHPS), ONS Longitudinal Study (ONS LS), the 2011 ONS Census
and Local Government election results from the University of Plymouth. We also
commissioned a specially-designed YouGov tracking survey and undertook four focus
groups, two in greater Birmingham and two in greater London. In general, we sought to
explain which factors accounted for White British attitudes to immigration, and were
associated with voting for the BNP or UKIP.
Ethnic Change and English Nationalism
The results of Gareth and my research suggest to us that English ethnic nationalism is
largely stimulated by ethnic change - a shift in the ethnic composition of the population which occurs through both immigration and differential birth rates between the White British
and other ethnic groups. The latter stems from minorities' younger age structure and formerly
higher birthrates as compared to the ethnic majority. English ethnic nationalism, like all
nationalism, seeks to align cultural and territorial boundaries. Growing diversity in England is
viewed as driving a wedge between the ethnic majority and 'its' homeland. The nationalist
imperative is to reduce this growing dissonance through some combination of slower
immigration, assimilation or, for extremists, repatriation of minorities. Majority ethnonationalism develops because the perceived continuity of 'English' cultural or physical traits
is thrown into question by change.
This perspective is common in the field of political demography. For instance, ethnic
conflicts in Northern Ireland, Lebanon, Israel-Palestine, Kosovo, Ivory Coast, Uganda and
Kenya have all been linked to fears, among regional or national ethnic majorities, of
impending ethnic shifts.2 These typically take one of two forms. Either 'advanced' groups of
settlers are viewed as 'invading' a minority group's region, i.e. Han Chinese settlers in Tibet,
Kikuyu migrants in the Kenyan Highlands, Javanese transmigrants in West Papua; or
minorities are perceived as growing, posing a threat to the majority, as with Catholics in
Northern Ireland after 1945 or Bengali Muslims in the state of Assam in India today. In many
ways, the situation in England is a milder version of the latter process. Note the link between
demography and politics. Democratisation means that numbers, i.e. demography, count for
votes. Hence differences in ethnic population growth rates are often politicised.3
Integration: Defusing the Demographic Challenge
In normal circumstances, integration is the main factor which counteracts the ethnic
nationalist anxiety wrought by ethnic change. Integration here refers to a loose set of
mechanisms: everything from minorities acquiring the English language to their
intermarrying or changing their identity to that of the majority. While all integrating
mechanisms play their role, the research Gareth Harris and I have conducted isolates two key
drivers. The first is simple habituation: here the English ethnic majority becomes used to the
presence of minorities even if they do not interact with them in intimate contexts. Minorities
gain a certain legitimacy - a 'right to be here' - in the eyes of the majority, while stereotypes
are challenged by reality. This takes the edge off former fears and hostility.
The second is assimilation. This denotes the full absorption of minorities into the
ethnic majority through intermarriage or boundary change. One can think of this as a kind of
dis-immigration. The two processes are not identical: the first mechanism involves the
absorption of individuals into the majority through intermarriage. Boundary change refers to
the ethnic majority widening its sense of self to include former outgroups. The best example
of boundary change in recent times has been the acceptance of Irish Catholics such as
Tommy Robinson of the EDL, or Jews such as Michael Howard, as White Englishmen. This
kind of change can suddenly usher large numbers of people inside the majority tent. The
feeling is mutual: majority acceptance begets minority assimilation. Today more children of
parents born in Ireland identify as White British than as Irish. Over three-fourths of those
specifying their religion as Jewish in the 2011 census identify as White British with less than
a quarter opting for 'White Other'.
Assimilation over generations is the typical route by which nonwhites join the ethnic
majority. The fact that around three-quarters of those of mixed Afro-Caribbean and white
background marry whites, and two-thirds of white-Asian mixed people do likewise, is
telling.4 This points in the direction of what Rogers Brubaker (2001) terms voluntary
assimilation. Blacks, for instance, have historically come to England and Wales via port cities
and were gradually absorbed into the white population.5 Only those whites who are aware of
family history, such as Cedric Barber, a white descendant of Samuel Johnson's freed slave
Francis Barber, are able to trace their black roots.6 In what follows, I consider what the data
tell us about the effect of ethnic change, habituation and assimilation.
Contact or Threat?: The Local and the National
Ethnic change affects a nation unevenly. It telescopes onto large cities but is
negligible in much of the countryside. In England, London and the West Midlands are the
most diverse regions, with cities more affected than market towns. Within cities, poorer builtup neighbourhoods with a high proportion of rental stock and population turnover tend to
change their ethnic composition fastest. The spatial diffusion of the minority and European
population refracts White British attitudes in disparate ways because for many people their
locale is an important means through which they imagine the nation. In a sense, the nation
becomes the local writ large.7 In the late nineteenth century, for instance, people in Hamburg
thought of the new Germany as a seafaring nation, projecting their Hanseatic memories onto
the nation. This was hardly the Bavarian view because Germans in the interior observed the
nation through a landlocked lens.8
The local is intimately tied to the national: the more of a 'local' one is, the more
national. Consider that people's threshold for ethnic diversity at local and national scales is
highly correlated. To understand this further, we commissioned a YouGov tracker survey of
1869 British adults in late July, 2013.9 Of those who responded, 1638, or 88 percent,
identified their ethnicity as White British. We asked, 'When do you think you would start to
feel uncomfortable about the number of people from ethnic minorities living in your
neighbourhood?' Answers covered an 11-item scale from trace numbers of ethnic minorities
to more than 75 percent minority.
We next applied the same scale to the national-level question, 'When do you think you
would start to feel uncomfortable about the number of people from ethnic minorities living in
Britain? For White British responses, the local comfort threshold predicts 60 percent of the
variation in national comfort threshold, an extremely powerful correlation. This means a one
unit move up the local comfort scale results in a 7/10 of a unit shift up the national comfort
scale. This neatly illustrates how local and national perceptions intertwine. Many craft their
ideal of the nation on the model of their locale: threats to the homogeneity of both are
perceived in similar ways.
Our work pays close attention to the impressive differences between the way White
Britons in diverse and homogeneous neighbourhoods view the immigration question. This is
tied, I maintain, to the way divergent local lenses refract the 'picture' of the nation that lives in
the mind's eye of White British people across England. Figure 1 presents the views of some
twenty thousand White British respondents on the question of whether immigration should be
increased, maintained or reduced. The proportion of White Britons opposing immigration in a
particular ward is presented on the vertical axis, the share of immigrants in the respondent's
ward of residence on the horizontal. Around 80 percent of White Britons felt immigration
should be reduced, with 60 percent calling for it to be reduced 'a lot'. Notice however that the
line declines from right to left, signfiying that whites in more diverse wards are more tolerant
of immigration. The pattern holds for the white working class as much as the white average,
though the contextual effects of diversity are greater among the upper-middle class than the
working class. The finding that more local diversity spawns a greater acceptance of diversity
is not obvious: we might expect opinion to be more hostile in areas affected by immigration.
Yet this finding holds up under rigorous statistical analysis, and has been replicated in
numerous studies of racism, far right voting and opposition to immigration.
Figure 1.
White British opposition to immigration,
by share of immigrants in ward
90.0%
85.0%
80.0%
75.0%
Whites
70.0%
White Working Class
65.0%
60.0%
55.0%
50.0%
15+%
Immigrants
10-15%
5-10%
2-5%
Less than 2%
immigrants
Source: Citizenship Surveys, 2010-11. Census 2011.
Our statistical analysis shows there are three reasons why opposition to immigration is
higher in areas with fewer immigrants. First, local whites have more opportunity to meet
minorities in immigrant areas, which helps produce the positive contact that ameliorates fears
and blunts stereotypes. This forms the basis of contact theory in social psychology, which is
discussed by Miles Hewstone in further detail in this issue. But even when we control for the
actual extent of inter-ethnic friendship and mixing, the share of minorities exerts a
prophylactic effect on white anti-immigration attitudes. This brings us to our second point.
That whites who live in diverse areas, even if they do not have meaningful contact with
minorities and immigrants, come to see them as legitimate. Finally, areas with large
immigrant populations tend to have a very transient white population, many of whom are
singles or renters. The transient atmosphere tends to reduce the extent to which local whites
perceive their neighbourhood - and by extension their nation - as 'turf' which is under threat.
Minorities or East European Immigrants?
In numerous opinion surveys, and in UKIP's campaign literature, European
immigration is a prominent concern while non-European immigration goes unmentioned. In
particular, many speak of the negative impact of European immigrants on the job prospects of
White British workers. But anti-prejudice norms operate to channel and limit the way people
express their anxieties.10 It is legitimate to oppose immigration but not diversity: therefore
arguing against European immigration on economic grounds is the most legitimate narrative.
Racist gaffes by UKIP activists such as William Henwood or Andre Lampitt greatly
embarassed the party because it prides itself on opposing white immigration from Europe
while supporting British ethnic minorities. An example of how the anti-prejudice norm
displaces cultural and ethnic concerns into the economic sphere comes from an older woman
in our Croydon focus group who commented that, coming in on the Croydon tramlink, 'I
might have been the only English person on that tram....I didn't like it...I could have been in a
foreign country.' Challenged by another participant, 'Why should that affect you that there's
minorities on the train,' the woman swiftly changed her narrative to a more acceptable,
economic, form of opposition to immigration: 'It doesn't affect me. It, um..I've got
grandchildren and children....I don't think things are going to get any better or easier for
them, to get work.'
This is not to say European immigration doesn't matter to the public. Indeed, the share
of post-2000 East European migrants in a ward, like the share of post-2000 non-European
immigrants, positively predicts opposition to immigration when controlling for confounding
factors. But when it comes to BNP voting, only local minority increase is significant. Ethnic
changes stimulate opposition to immigration but a higher level of established minorities and
'white others' (European-origin whites), namely those who arrived in the UK prior to 2000,
correlates with reduced opposition to immigration. Europeans and minorities begin to diverge
only when we examine UK-born generations. In our quantitative analysis, the rate of increase
in UK-born minorities is associated with increased opposition to immigration whereas a rise
in the share of UK-born Europeans is not. More than this, the extent to which minorities
speak English as their native tongue, are employed, homeowners, possess UK passports or
identify as British seems to have no impact on majority perceptions.
Census and survey results show that many children of Southern and Eastern European
migrants identify as White British whereas few of non-European background do. This is
probably because, with the exception of some of those of mixed-race, visible minorities are
not accepted by the majority as fully 'one of us'. European assimilation, by contrast, is brisk.
Even in the period 2001-11, some 7 percent of those who ticked the 'White Other' box in
2001 called themselves White British in 2011, net of those going the other way. Among those
who listed their parents' birthplace as 'Mediterranean Commonwealth' on the census in 1971,
46 percent considered themselves members of the dominant White British group by 2011.
This is a pace of assimilation in line with that of the Irish.
All told, our research tells us that White British people habituate to both East
European and non-European immigrants after a ten-year period. Assimilation, however,
seems to occur more rapidly among European-origin than minority newcomers. In short, for
the ethnic majority, the immigration issue is not, as commentators such as Sunder Katwala
sometimes argues, reducible to an integration problem - except in a longer-term sense.
Acceptance of immigrants is both faster than Katwala allows, in the sense that habituation
transpires over a one-decade timespan, and slower, in that cultural and economic integration
has a limited impact on popular attitudes while minorities assimilate into the White British
group only through the slower process of intermarriage.
Ethnic Change and Ethnic Threat
From our previous discussion of contact theory, one might imagine that more local
minorities translates into more positive majority attitudes to immigration. But this is only true
of contexts smaller than that of the ward (population around 7,500). In larger geographies,
more minorities is associated with more hostile white opinion and voting. Thus the more
diverse a White Briton's Local Authority (population 100,000-200,000), the more a person
opposes immigration and the more likely they are to vote for the British National Party. The
'optimum' configuration for hostility to immigration is thus a white, close-knit ward
embedded within a diverse Local Authority. Some Scandinavian work refers to the 'halo
effect' of strong anti-immigrant hostility in the white outlying areas that ring diverse cities.11
In this way we can understand how the most fertile terrain for the National Front in London
in the 1970s was in Inner East London whereas today the BNP polls best in Outer London
boroughs like Barking and Dagenham, or in adjoining areas of Essex and the Thames
Gateway. In Antwerp, the Vlaams Belang has similarly seen its core areas of support move
outwards as former strongholds acquired a more diverse population.
How do we make sense of the dual pattern of threat from diversity in one's wider area
combining with threat reduction from local diversity? One clue comes from survey data.
When asked whether immigration is a problem in their local area, few reply in the
affirmative. When the question is phrased as a 'problem in the nation', most answer yes. In
fact the gap between perceiving immigration as a local and national problem is, as Bobby
Duffy and his colleagues show, a full 50 points. Duffy's respondents explained the
discrepancy by claiming to have observed immigration as a national problem when visiting
other towns and cities or through the media. When pressed on the local-national discrepancy
regarding immigration in our focus groups, most people cited media stories about immigrants
boarding lorries in Calais or second-hand information from friends about problems in cities
such as Leicester. As Charlie Leddy-Owen shows in this volume, respondents are skeptical of
the tabloid media, yet simultaneously cite it as a reliable source of information on migrants.
The presence of significant diversity in one's city or Local Authority adds to threat
perceptions because of the sense immigrants may soon introduce large-scale change into
one's locale. Dan Hopkins notes that people tend to be most fearful of nuclear power plants
when they live close, but not too close, to a plant.12 Those far away are untroubled as are
those who live next to the plants and understand their dangers more intimately. Likewise the
'halo effect': the fact opposition to immigration is greatest when immigrants are close, but not
too close, accounts for the differing effects of minority population share on white perceptions
of immigration (and BNP voting) at the ward and Local Authority levels. By contrast, whites
in diverse wards nested in white Local Authorities would tend to be most tolerant because
minorities are perceived as locally established yet are too small a proportion to be threatening
to those in the wider area.
Local Ethnic Change and the White British Response
A second caveat to the idea that diversity changes white attitudes is that white toleration is
driven by levels of diversity but reduced by changes in diversity. Areas such as Brixton in
London have a tradition of being diverse, thus local whites are tolerant of immigration. Yet in
Barking and Dagenham, diversity had been minimal until the 2000s whereupon it took off.
Rapid ethnic change, especially in places with limited experience of prior diversity, tends to
be associated with radicalised white opinion and elevated far right voting. Figure 2 plots
wards, labelled by the Local Authority in which they are nested, with the BNP share of the
vote in local elections of 2010-12 on the vertical axis and a measure of the speed of minority
growth on the horizontal.
Figure 2.
Sandwell
Bexley
Ealing
Redbridge
Thurrock
Havering
Hounslow
Newcastle
upon
Tyne
Barking and
Dagenham
Bexley
Barking
and
Dagenham
Pendle
Thurrock
Burnley
Calderdale
Broxbourne
Sandwell
Southwark
Thurrock
Sandwell
Tower
Hamlets
Croydon
Leeds
SandwellBurnley
Leeds
Kirklees
Thurrock
Kirklees
Greenwich
Greenwich
Stoke-on-Trent
Stoke-on-Trent
Amber
Valley
Sandwell
Croydon
Burnley
Calderdale
Havering
SandwellPendle
Barking
and Dagenham
Rotherham
Burnley
Greenwich
Redbridge
Birmingham
Rochford
Hounslow
Sandwell
Nuneaton
and
Bedworth
Kirklees
Epping
Forest
Hounslow
Calderdale
Hillingdon
Rotherham
Sandwell
Barnsley
Manchester
Hounslow
MertonBexley
Manchester
Bradford
Burnley
Hillingdon
RotherhamKirklees
Burnley
Wakefield
Bradford
Enfield
Hillingdon
Thurrock
Manchester
Greenwich
Rotherham
Bradford
Epping
Forest
Wakefield
Crawley
Crawley
Bexley
Epping
Forest
Bexley
Sandwell
Amber
Valley
Barnsley
Sheffield
Newcastle
upon TyneMerton Redbridge
Brent
Hertsmere
Stoke-on-Trent
Birmingham
Kirklees
Sandwell
Sandwell
Sheffield
Solihull
Tameside
Sandwell
Tameside
Havering
Bradford
Barnsley
Hounslow
Burnley
Sandwell
Rotherham
Solihull
Bradford
Enfield
BexleyMerton
Hounslow
Pendle
Stoke-on-Trent
Leeds
Leeds
Pendle
Amber
ValleyWolverhampton
Enfield
Crawley
Barnsley
Greenwich
Rotherham
Sandwell
Sandwell
Bradford
Barnsley
Bexley
Sutton
Sheffield
Bexley
Manchester
EalingTower
Worcester
Tameside
Birmingham
Broxbourne
Bury
Hamlets
Kirklees
Sandwell
Dudley
South
Nuneaton
Tameside
Tyneside
and
Bedworth
Thurrock
Leeds
Barnsley
Thurrock
Nuneaton
Birmingham
Salford
andHamlets
Bedworth
Amber
Rotherham
Valley
Bradford
Broxbourne
Birmingham
Barnsley
Solihull
Calderdale
Sandwell
Tameside
Greenwich
Kirklees
Wigan
Tower
Barnsley
Greenwich
Wakefield
Nuneaton
Wakefield
and Bedworth
Newcastle
Tyne
Barnsley
Pendle
Barnsley
Redditch
Leedsupon
Salford
Doncaster
Rochdale
Derby
Sandwell
Leeds Birmingham
Bradford
Wigan
Pendle
Wolverhampton
Bexley
Walsall
Barnsley
Newcastle
upon
Tyne
Burnley
South
Tyneside
Nuneaton
and
Bedworth
Bexley
Basildon
Birmingham
Sutton
Tameside
Wakefield
Sunderland
Nuneaton
and
Bedworth
Bexley
Wigan
Coventry
Bexley
Bury
Doncaster
Tameside
Wigan
Nuneaton
and
Bedworth
Rotherham
Cannock
Chase
Basildon
Birmingham
Kirklees
Wolverhampton
Birmingham
South
Tyneside
Worcester
Salford
Merton
Bolton
Crawley
Bromley
Derby
Stoke-on-Trent
Barnsley
Leeds
Sandwell
Leeds
Southend-on-Sea
Bexley
Salford
Burnley
Havering
Sheffield
Bury
Bexley
Greenwich
Solihull
Wakefield
Nuneaton
Bexley
and
Bedworth
Barnsley
Thurrock
Doncaster
Rochdale
Wakefield
Wigan
Bradford
Hertsmere
Rochford
Wigan
Barnsley
Wakefield
Leeds
Nuneaton
and
Bedworth
Sandwell
Tower Hamlets
North
Tyneside
Birmingham
Tameside
Thurrock
Derby
Greenwich
Rotherham
Wakefield
Coventry
Redbridge
Bexley
Walsall
Sandwell
Worcester
Burnley
Sandwell
Worcester
Manchester
Newcastle upon Tyne
Basildon
Birmingham
Coventry
Doncaster
Worcester
Merton
Kirklees
Bromley
Pendle
Leeds
Birmingham
Sheffield
Barnsley
Tameside
Greenwich
Sutton
Richmond
upon Thames
South
Tyneside
Basildon
Hastings
Wigan
Warrington
Greenwich
Hartlepool
Barnsley
Barnsley
Rotherham
Nuneaton
and
Bedworth
South
Barnsley
Tyneside
Rochdale
Leeds
Wakefield
Coventry
Birmingham
Birmingham
Newcastle
Wakefield
upon
Tyne
Wakefield
Sheffield
Kirklees
Coventry
Coventry
Broxbourne
Basildon
Sandwell
Amber
Valley
Swindon
Greenwich
Islington
Wigan
Stockport
Birmingham
Pendle
Oldham
Thurrock
Solihull
Bury
South
Tyneside
Crawley
Bristol,
City
of
Barnsley
Bassetlaw
Solihull
Bexley
Wigan
South
Tyneside
Birmingham
Hartlepool
Stockport
Sutton
Milton
Keynes
Dudley
Thurrock
Leeds
Sheffield
Gateshead
Bristol,
City
of
Wakefield
Basildon
Southend-on-Sea
Sunderland
Birmingham
Basildon
Tameside
Sunderland
Nuneaton
Wakefield
Gateshead
and
Bedworth
Nuneaton
and
Bedworth
Harrogate
Tameside
Crawley
Tameside
Sheffield
Bury
Doncaster
Carlisle
Bristol,
Tameside
City
of
Sunderland
Sunderland
Newcastle
Leeds
upon
Tyne
Croydon
Manchester
Basingstoke
and
Deane
Barnsley
Amber
Valley
Newcastle
Tyne
Birmingham
Birmingham
Croydon
St.
Helens
Bexley
Thurrock
Kirklees
Enfield
Wigan
Wakefield
Calderdale
Salford
Birmingham
Derby
Wigan
Salford
Swindon
Birmingham
South
Halton
Tyneside
Kirklees
Wakefield
Basildon
Hyndburn
Bury
Bristol,
City
of upon
South
Leeds
Tyneside
Cannock
Wigan
Chase
Camden
Thurrock
Thurrock
Gateshead
Hastings
Bexley
Bristol,
City
of
Liverpool
Coventry
Sunderland
Gateshead
Bolton
South
South
Wigan
Tyneside
Tyneside
Southend-on-Sea
Gateshead
Newcastle
Sunderland
upon
Tyne
Bolton
Wakefield
Gateshead
Swindon
Salford
Tameside
Gateshead
Coventry
Hartlepool
South
Tyneside
Manchester
Basildon
Solihull
Stockport
Reigate
and
Banstead
Sheffield
Bexley
Epping
Forest
Rochford
Newcastle
Coventry
upon
Tyne
Calderdale
Sunderland
Worcester
Croydon
South
Newcastle
Lincoln
Tyneside
upon
Tyne
Sheffield
Barnet
Leeds
Newcastle
Newcastle
Leeds
upon
upon
Tyne
Tyne
Carlisle
Coventry
Gateshead
Swindon
Coventry
Kirklees
Kirklees
Bury
Richmond
upon
Thames
Kirklees
Southend-on-Sea
South
Bromley
Wakefield
St.
Tyneside
Helens
Newcastle
Rochdale
Solihull
upon
Tyne
Wigan
North
Liverpool
St.
Helens
Bristol, City of
Stockport
Gateshead
Basildon
Wirral
Kingston
C
arlisle
Southend-on-Sea
upon
Birmingham
Gateshead
Hull,
City
of
Sheffield
Solihull
Wakefield
Kingston
upon
Croydon
Hull,
City
of
South
Wigan
Tyneside
Kirklees
Wigan
North
Sheffield
East
Lincolnshire
Bury
Ipswich
St.
Gateshead
Helens
Birmingham
Birmingham
Bristol,
City
Southend-on-Sea
of
Gateshead
Bolton
Harrogate
Salford
Wandsworth
Wigan
Bury
Newcastle
Bury
Liverpool
upon
Tyne
Knowsley
Carlisle
Sunderland
Liverpool
Harrogate
Southend-on-Sea
Coventry
Leeds
Bristol,
City
of
Wolverhampton
Solihull
Stockport
Lincoln
North
Hertfordshire
Gateshead
Birmingham
Newcastle upon Tyne
Leeds
Newcastle
Carlisle
upon
Tyne
St.
Helens
Stockport
Kirklees
Hastings
Southend-on-Sea
Derby
Southend-on-Sea
Liverpool
Leeds
Plymouth
St.
Sheffield
Helens
Wirral
Basildon
Salford
Bristol,
City
ofRedbridge
Swindon
Stevenage
Solihull
Carlisle
Birmingham
Tameside
Wigan
Lincoln
Kirklees
Gateshead
Sunderland
Southend-on-Sea
Leeds
Kirklees
Solihull
Carlisle
Barnsley
North
Hertfordshire
Carlisle
Hartlepool
Wirral
Croydon
Birmingham
Camden
Solihull
Manchester
Southend-on-Sea
Reigate
and Birmingham
Banstead
Liverpool
Dudley
St.
Helens
Sunderland
Carlisle
Southend-on-Sea
Sefton
Bristol,
City of
Liverpool
Rochford
Stockport
Leeds
Kirklees
Sefton
Bristol,
City
of
Basildon
Lincoln
Southend-on-Sea
North
Hertfordshire
Barrow-in-Furness
North
Tyneside
Leeds
Sunderland
Newcastle
upon
Tyne
Stockport
Liverpool
Coventry
Southend-on-Sea
Birmingham
Newcastle
upon Tyne
Leeds
Sefton
Leeds
Camden
Solihull
Gateshead
Wigan
Leeds
Southend-on-Sea
Sefton
Solihull
North
Tyneside
Bristol,
City
of
South
Basildon
Tyneside
Stockport
Southend-on-Sea
Newcastle
upon
Tyne
St.
Wirral
Wirral
Helens
Birmingham
Leeds
Basildon
Leeds
Leeds
Southend-on-Sea
Sefton
Exeter
Sefton
Sefton
Sefton
Leeds
Hastings
Colchester
Leeds
Newcastle
upon
Tyne
Solihull
Sefton
Newcastle
upon Tyne
Newcastle
upon
Tyne
Barking and Dagenham
0
% BNP
British Only) 30
10Vote (White20
BNP Vote and Ethnic Shift in Local Elections, 2010-12
0
1000
2000
3000
% White British 2001 x Change in Minority Share '01-11
wards
4000
Fitted values
Source: Calculated from Election data from Plymouth Elections Centre Data13; Census 2011.
The upward slope and tight bunching of dots around the line indicates that BNP support is
strongest in wards that were relatively White British in 2001 but experienced a fast increase
in ethnic minority share during the 2000s. For example, a number of wards in Outer London
or Essex boroughs such as Redbridge, Barking and Dagenham or Thurrock changed quickly
and saw high BNP support.
The line is considerably less steep for UKIP: in the local elections of 2010-12, UKIP
supporters were much more likely to live in whiter, rural constituencies than BNP voters.
This reflects the fact that UKIP is concerned with a broader suite of issues than the BNP,
encompassing, in particular, European integration. When one factors in the age and class
profile of UKIP voters, their views on immigration are less distinct from the average than
those of the BNP, whose voters are younger and more urban. Our data are for local elections
of 2010-12, however, and there is some evidence from the 2014 British Election Study (see
Matt Goodwin and Rob Ford's contribution) that UKIP's support profile may have moved in a
more working-class, northern, immigration-focused direction since. The increase in minority
share is associated in a less direct way with UKIP support insofar as increases in minority
share at ward level correlate with a higher share of White British people identifying their
national identity as English rather than British. Even more so than the BNP, UKIP-supporting
wards in the 2010-12 local elections are located in Local Authorities where a disproportionate
share of White Britons identify as English rather than British. Moreover, English national
identity is one of the few solid individual-level predictors of UKIP support in our 2009-12
Understanding Society survey data.
Residential Integration
The BNP did well in places with fast ethnic change, like Barking, but also polled
strongly in Burnley, which hasn't experienced the same rate of change. What might explain
this? Here segregation plays a role, which gets back to our point about the divergent effects
of ward and Local Authority diversity on white opinion. Quantitative analysis shows the
White British Isolation Index, a measure of how segregated white Britons are from
minorities, is associated with more far right voting and less favourable views of immigration.
By contrast, residential integration helps break down peoples' suspicion of immigrants.
Hostility to immigrants in the abstract is adulterated somewhat by the humanising
influence of concrete immigrants and minorities. In the diverse Lozells area of Birmingham
one lady in our focus groups related: 'I'm from Hereford...[The diversity of Lozells] was a bit
of a shock to start with...being here has made me realise there are some lovely ethnic
minorities around.' A woman from our Croydon focus group spoke of the bad publicity that
Kosovan refugees received in 1999. Befriending a Kosovar interpreter she came to realise, 'a
lot of things that was [sic] in the media didn't happen.' Contact also seems to be linked to
mixed partnerships, or to a greater incidence of having mixed-race relatives. In our Croydon
focus group, two of fifteen present had mixed-race childen, another's partner had a mixedrace daughter, and others had dated minorities or immigrants. Thus the longer-term effects of
residential integration are probably more important than the short-term ones.
We noted that higher local levels of minorities reduces opposition to immigration. Yet
minorities are extremely unevenly spread around England. A fifth of England's minorities
live in just 166 of England's 8850 wards: it takes over 6700 homogeneous wards to amass
the same number. An improved distribution of minorities could reduce threat and improve
contact to some extent - though the spread of minorities may also spread threat in the 'halo'
rings beyond new minority settlement areas. The net effect of residential integration on
majority attitudes is therefore moderate in the short term and cannot, on its own, offer a
panacea for the immigration problem. However, in the long run, integration smooths the path
to assimilation, which does reduce ethnic tensions and opposition to immigration.
An Inclusive Britishness: Can the majority's identity be broadened by the state?
Sensing the advantages of majority-group boundary extension, liberal policymakers
who support immigration sometimes place their faith in engineering an inclusive identity
from above. From a historical perspective, I would argue such attempts are unlikely to
succeed because national identity today is a complex system that emerges from individuals,
organisations, media outlets and other variegated aspects of a complex society. The days
when a central authority could implement a new version of national identity from on high,
akin to the French Third Republic post-1870, are behind us as the media landscape has
become fragmented, with reduced trust in government. As Gordon Brown's Britishness
campaign revealed, the state is but one voice in a many-sided conversation whose balance of
opinion can readily frustrate that of the state.
National identities are contested by different ideological fragments in society. A proimmigrant version of English national identity exists, championed by liberals of the right,
such as the Financial Times, or by left-liberals at the Guardian. Both encourage open
immigration and a fluid, forward-looking national identity. They endorse a variant of national
identity at odds with ethnic interpretations of Englishness and would applaud the prose of
Daniel Defoe some three centuries ago:
Thus from a mixture of all kinds began/That het’rogeneous thing, an
Englishman...From whence a mongrel half-bred race there came...In whose hot veins
new mixtures quickly ran...Infus’d betwixt a Saxon and a Dane/While their rank
daughters, to their parents just/Receiv’d all nations with promiscuous lust. (Daniel
Defoe, 'The True-born Englishman', 1701)
Liberals of left and right concur with Heraclitus in viewing England and Britain as an everchanging flow. Put your foot in the stream twice and you experience different Englands.
Against this view, conservatives of the right and left (i.e. Old Labour) view the nation as a
timeless river. England in all its particularity endures, outliving any particular generation.
Interpretation is important but the historical record also matters: nobody could
convince today's English they are actually Frenchmen. In North America, liberals have,
especially since the 1960s, successfully contested ethnic perspectives on the nation. An
obvious indicator is the elimination of the 'national origins' quotas in American immigration
policy with the Hart-Celler Act of 1965. Another is the emphasis in both Canada and the US
on the 'nation of immigrants' and 'multicultural' narratives - evident in the polyglot
iconography and bilingual Spanish-English signs in many American airports. This view of
the nation remains contested, not least by the Tea Party, whose media mouthpiece Glenn
Beck railed against the multicultural 2014 Coke Super Bowl commercial which featured
America the Beautiful being sung by a multiracial cast in different languages. However, this
reframing, which was only possible due to the conjunction of the Civil Rights movement and
the liberal culturequake of the sixties, was the product of a specific historical moment. This
window has now closed, and the sixties did not produce the same recasting of national
identity in England as it did in post-Kennedy America.
The antiquity of Anglo-Saxon settlement in England (560 as opposed to 1607-20 for
the US) and the more limited experience of immigration prior to 1948 help tip the balance
toward nationalists who argue that a 'native' English ethnic group had formed and become
established, retaining its broad characteristics over centuries. Most ethnic majorities in
western Europe have been using their group's proper name since before 1500, thus claims to
majority-group indigenousness are more difficult for liberals to unseat than in the New
World.14 This makes it harder for cultural liberals to reframe European ethnonationalism in
multicultural or civic terms. The recent European experience, in which far right parties
speaking the language of indigenous nationalism have gained significant ground is therefore
more relevant than in North America, where such movements only exist on the fringe of
political life.
Individualism: Escaping the Chains of Ethnicity and Nationhood?
Ted Cantle (see this volume) claims individualism loosens the bonds of ethnic
nationhood, inaugurating a post-ethnic, postnational sensibility. Greater individualism
attenuates people's attachments to ascribed identities such as ethnic group and nation.
Modernity has brought increased wealth and education, greater physical separation of people,
the rise of nuclear and single living, mass literacy and private entertainment technology. All
of which empower individuals to slough off identities tied to communities of history. When
people become more individualistic, they no longer rely on the symbolic continuity of the
nation as a vessel through which the story of their lives unfolds. Instead, they come to narrate
their existence as a story of their own achievements and lifestyle choices, or attach
themselves to social categories like generation or subculture which tend to be time-bound. As
a result, individualism weakens communal bonds and identities.
The big individualising changes took place in the late twentieth century. This is to
some degree an unfinished revolution: the continuation of some individualistic trends secularism, university education - should be expected. Against this, Britons move less now
than they did a decade or two age, the population is aging and the trend toward higher divorce
rates has plateaued, portending a flattening or reversal of post-60s individualism. Young
people are more individualistic than the old, but as people age, their views on certain issues
become more conservative. We know that support for the Tories, for instance, increases as a
function of age even when all other factors are considered.15 This is less the case for social
issues such as homosexuality and religion, but has been true of European identity and seems
to be the case for immigration, as Bobby Duffy and his collaborators show in the impressive
report which underpins his article in this issue.
Among our London and Birmingham focus group respondents there was a sense that
tolerance was only partly generational - that this could dissipate with age as individuals
encounter adversity. A younger Lozells (Birmingham) man agreed that '[youth are] more
tolerant cause they're brought up around it' but added 'as you get older you...see for yourself
what's going on' and grow more skeptical. A Lozells woman admitted that when she entered
College all were tolerant of diversity and when one is young 'you don't have any hate in you.'
However, as she has got older, her views have grown more negative on immigration. This
would seem to be borne out in data on immigration by generation which shows the 60s-era
Baby Boomers experiencing a shift in later life toward more negative attitudes (see Duffy's
contribution in this issue). Generation X (born 1966-79) and Millennials (1980-2000) exhibit
a more tolerant trend, but it is difficult to extrapolate from this toward a more tolerant future.
To wit, individualism is unlikely to serve as a powerful solvent of English ethno-nationalism
in the present-day.
Conclusion
I argue that in most times and places, ethno-demographic change breeds an anti-immigration
response among the majority. This has certainly been the case in the United States, England
and Scotland from the late nineteenth century until the late twentieth century. Change tends
to pry apart the alignment of ethnicity and territory cherished by ethnic nationalists. The
popular English ethnic nationalism of UKIP, which taps a much broader vein of nationalism
amongst the majority population, seeks to reduce this dissonance by reducing immigration,
slowing the rate of ethnic change. Individualism and major ideological shifts can alter this
equation and usher in a mood of tolerance for high immigration, as in North America since
the 1960s, or, more recently, in Germany.
However the case of England is more in line with the historical norm than these
tolerant exceptions. Accordingly, integration offers the best chance for a reduction in popular
anti-immigration feeling which has powered UKIP's rise. This said, integration takes time.
Most of its effects are felt over decades or generations as the ethnic majority becomes
habituated to the presence of local minorities and the children of minorities are absorbed into
the majority population. Minority birthrates have fallen sharply in the past few decades,
intermarriage is on the rise and the children of European immigrants tend to assimilate into
the ethnic majority. Immigration from outside Europe has fallen. These forces are likely to
slow the pace of ethnic change and quicken that of integration. Nevertheless, such changes
cannot occur within an election cycle, thus the immigration issue will be with us for some
time, providing fuel for the UKIP juggernaut. Mainstream politicians and policymakers are
advised to adopt a more realistic approach to the issue while taking steps to facilitate the
integration of minorities with the English ethnic majority.
1
ESRC ref# ES/K003895/1.
Kaufmann, Eric. 2011. "The Demography of Ethnic Conflict." Journal of Ethnopolitics 10 (3-4):367-8.
3
Goldstone, Jack A., Eric Kaufmann and Monica Duffy Toft, ed. 2012. Political Demography: How Population
Changes are Reshaping National Politics and International Security, edited by J. A. Goldstone, E. Kaufmann
and M. Toft. Oxford: Oxford University Press.
4
Voas, D. 2009. "The Maintenance and Transformation of Ethnicity: Evidence on Mixed Partnerships in
Britain." Journal of Ethnic and Migration Studies 35 (9):1497-513.
5
Holmes, Colin. 1988. John Bull's Island: Immigration and British Society, 1871-1971. London: MacMillan
Education Publishers.
2
6
< http://www.bbc.co.uk/videonation/articles/s/stoke_abolition2.shtml>
Kaufmann, Eric. 2008. "The Lenses of Nationhood: An Optical Model of Identity." Nations & Nationalism 14
(3):449-77.
8
Umbach, M. 2006. "Germany." In What is a Nation?: Europe, 1789-1914, ed. T. Baycroft and M. Hewitson.
Oxford: Oxford University Press.
9
http://d25d2506sfb94s.cloudfront.net/cumulus_uploads/document/kf5d231qce/YG-Archive-ESRC-DemosBirkbeck-results-300713.pdf
10
Blinder, S., R. Ford, and E. Ivarsflaten. 2013. "The Better Angels of Our Nature: How the Antiprejudice
Norm Affects Policy and Party Preferences in Great Britain and Germany." American Journal of Political
Science 57 (4):841-57.
11
Rydgren, J., and P. Ruth. 2013. "Contextual explanations of radical right-wing support in Sweden:
socioeconomic marginalization, group threat, and the halo effect." Ethnic and Racial Studies 36 (4):711-28.
12
Hopkins, A. 2013. "Misplaced: The Limits of Contextual Influence on Americans' Political Attitudes " In
American Political Science Association. Chicago, IL.
13
http://www1.plymouth.ac.uk/research/ceres/TEC/thecentre/Pages/default.aspx
14
Kaufmann, Eric. 2014. "Land, History or Modernization: explaining ethnic fractionalization." Ethnic & Racial
Studies, forthcoming.
15
Evans, Geoffrey and James Tilley. 2014. "Ageing and generational effects on vote choice: Combining
cross-sectional and panel data to estimate APC effects," Electoral Studies, 33(14), 19-27.
7